Biology subjects

Horenko, I.

Publications and source records attributed to Horenko, I..

4 recordsLinked to original sources

bioRxiv2023

Spiking neural networks provide accurate, efficient and robust models for whisker stimulus classification and allow for inter-individual generalization

With the help of high-performance computing, we benchmarked a selection of machine learning classification algorithms on the tasks of whisker stimulus detection, stimulus classification and behavior prediction based on electrophysiological recordings of layer-resolved local field potentials from the barrel cortex of awake mice. Machine learning models capable of accurately analyzing and interpreting the neuronal activity of awake animals during a behavioral experiment are promising for neural prostheses aimed at restoring a certain functionality of the brain for patients suffering from a severe brain injury. The liquid state machine, a highly efficient spiking neural network classifier that was designed for implementation on neuromorphic hardware, achieved the same level of accuracy compared to the other classifiers included in our benchmark study. Based on application scenarios related to the barrel cortex and relevant for neuroprosthetics, we show that the liquid state machine is able to find patterns in the recordings that are not only highly predictive but, more importantly, generalizable to data from individuals not used in the model training process. The generalizability of such models makes it possible to train a model on data obtained from one or more individuals without any brain lesion and transfer this model to a prosthesis required by the patient. Author SummaryA neural prosthesis is a computationally driven device that restores the functionality of a damaged brain region for locked-in patients suffering from the aftereffects of a brain injury or severe stroke. As such devices are chronically implanted, they rely on small, low-powered microchips with limited computational resources. Based on recordings describing the neural activity of awake mice, we show that spiking neural networks, which are especially designed for microchips, are able to provide accurate classification models in application scenarios relevant in neuroprosthetics. Furthermore, models were generalizable across mice, corroborating that it will be possible to train a model on recordings from healthy individuals and transfer it to the patients prosthesis.

neuroscience↗

bioRxiv2021

Quality-preserving low-cost probabilistic 3D denoising with applications to Computed Tomography

We propose a pipeline for a synthetic generation of personalized Computer Tomography (CT) images, with a radiation exposure evaluation and a lifetime attributable risk (LAR) assessment. We perform a patient-specific performance evaluation for a broad range of denoising algorithms (including the most popular Deep Learning denoising approaches, wavelets-based methods, methods based on Mumford-Shah denoising etc.), focusing both on accessing the capability to reduce the patient-specific CT-induced LAR and on computational cost scalability. We introduce a parallel probabilistic Mumford-Shah denoising model (PMS), showing that it markedly-outperforms the compared common denoising methods in denoising quality and cost scaling. In particular, we show that it allows an approximately 22-fold robust patient-specific LAR reduction for infants and a 10-fold LAR reduction for adults. Using a normal laptop the proposed algorithm for PMS allows a cheap and robust (with the Multiscale Structural Similartity index > 90%) denoising of very large 2D videos and 3D images (with over 107 voxels) that are subject to ultra-strong Gaussian and various non-Gaussian noises, also for Signal-to-Noise Ratios much below 1.0. The code is provided for open access. One-sentence summaryProbabilisitc formulation of Mumford-Shah principle (PMS) allows a cheap quality-preserving denoising of ultra-noisy 3D images and 2D videos.

bioinformatics↗

bioRxiv2019

Low-cost scalable discretization, prediction and feature selection for complex systems

Finding reliable discrete approximations of complex systems is a key prerequisite when applying many of the most popular modeling tools. Common discretization approaches (for example, the very popular K-means clustering) are crucially limited in terms of quality and cost. We introduce a low-cost improved-quality Scalable Probabilistic Approximation (SPA) algorithm, allowing for simultaneous data-driven optimal discretization, feature selection and prediction. Cross-validated applications of SPA to a range of large realistic data classification and prediction problems reveal drastic cost and performance improvements. For example, SPA allows the unsupervised next-day surface temperature predictions for Europe with the mean crossvalidated one-day prediction error of 0.75{degrees}C on a common PC (being around 40% better in terms of errors and five to six orders-of-magnitude cheaper than the next-day surface temperature predictions calculated on supercomputers and provided by the weather services).\n\nOne Sentence SummaryIntroduced computational tool allows obtaining drastic cost and quality gains for a broad range of science applications.

bioinformatics↗

bioRxiv2018

Imputation of posterior linkage probability relations reveals a significant influence of structural 3D constraints on linkage disequilibrium

Genetic association studies have become increasingly important in unraveling the genetics of diseases or complex traits. Despite their value for modern genetics, conflicting conclusions often arise through the difficulty of confirming and replicating experimental results. We argue that this problem is largely based on the application of statistical relation measures that are not appropriate for genomic data analysis and demonstrate that the standard measures used for Genome-wide association studies or genomics linkage analysis bear a statistic bias. This may come from the violation of underlying assumptions (such as independence or stationarity) as well as from other conceptual limitations in the measures or relations, such as missing invariance with respect to coding or the inability to reflect latent factors. Attempts to introduce unbiased relation measures that avoid these limitations are usually computationally expensive and do not scale for large data sizes being typical for genomics applications.\n\nTo tackle these problems, we propose a straightforwardly computable relation measure called Linkage Probability (LP). This measure provides the posterior probability of a relation between two categorical data sets and considers potential biases from latent variables. We compare several aspects of popular relation measures through an illustrative example and human genomics data. We demonstrate that the application of LP to the analysis of Single Nucleotide Polymorphisms (SNP) reveals latent 3D steric effects within 1D SNP data, that approximate to chromatin loops captured by high resolution Hi-C maps.

bioinformatics↗